In recent years, research on next-generation mobile communication systems has been actively conducted. In the research on such next-generation mobile communication systems, a cellular system which repeatedly uses one frequency has been proposed as a system for increasing the frequency use efficiency of the system. In this cellular system, each cell uses the same frequency band, thereby enabling each cell to use all bands allocated to the system.
As a communication scheme to be used by downlink, an OFDMA (Orthogonal Frequency Division Multiple Access) scheme is the most prominent candidate. The downlink refers to communication from a base station device to a mobile station. The OFDMA scheme is a system which performs communication by flexibly allocating radio resources to a plurality of mobile terminal devices using an OFDM signal to be communicated with a different modulation scheme for information data in response to a reception situation. Modulation schemes are 64QAM (64-ary Quadrature Amplitude Modulation), BPSK (Binary Phase Shift Keying), and the like. The radio resources are constituted by a time axis and a frequency axis.
In general, even when frequency selectivity fading occurs in the entire transmission band for an OFDM signal, a delay wave of a propagation channel may be treated as flat fading in view of a sub-carrier unit. This is because each sub-carrier is a narrow band.
Since control may be independently performed in the sub-carrier unit, inter-code interference caused by the delay wave does not exist and equalization is not necessary. Thus, the mobile station may detect a received signal of each sub-carrier in a reception situation as it is. Accordingly, the mobile station may appropriately control the quality of transmission when an adaptive modulation scheme is used to allocate an appropriate modulation scheme in response to the reception situation.
In this case, a PAPR (Peak to Average Power Ratio) may be greatly raised to use an OFDM signal. The occurrence of high peak power is not a large problem in downlink communication which has a relative margin for amplifying transmission power. However, the occurrence of high peak power is a fatal problem in uplink for which no margin exists for amplifying transmission power. The uplink refers to communication from the mobile station to the base station device.
Thus, in the uplink, it is desirable to use a communication scheme based on a single carrier scheme having a low PAPR.
However, when the communication scheme based on the single carrier scheme is used, a sub-carrier may not be independently processed as in the OFDM scheme. Thus, since inter-code interference caused by a delay wave may not be suppressed, an adaptive equalization technique is necessary to suppress inter-code interference of a reception signal.
As an adaptive equalization technique, there has been proposed a frequency domain equalization technique (FDE: Frequency Domain Equalization) in which equalization is possible by one product operation in frequency domain by adding a CP (Cyclic Prefix) for the purpose of maintaining periodicity for a time signal into which a plurality of transmission signals has been blocked even under a multipath fading environment and removing the CP by a reception side (Non-Patent Document 1). Since an FFT unit becomes a block, the blocked transmission signals are referred to as an FFT block.
In the frequency domain equalization technique (FDE), a type of ZF (Zero Forcing) is known which restores a signal component by multiplying an inverse characteristic of a frequency response of a propagation channel based on the fact that the convolution of an impulse response on the time axis is a multiplication of the frequency domain. However, since thermal noise is reliably added to a received signal inside the mobile station when the equalization is performed by a reception side, the inverse characteristic of the propagation channel is multiplied even in the terms of noise when the inverse characteristic of the propagation channel is multiplied. As a result, there is a problem in that a transmission characteristic thereof is worse than an actual transmission characteristic since the noise is colored and emphasized.
For the purpose of suppressing noise emphasis, MMSE-FDE based on the MMSE (Minimum Mean Square Error) criterion has been proposed to minimize the square of an error between a signal after equalization and a transmission signal. A tap matrix of MMSE-FDE is expressed by Equation (1).WH=(ΞΞH+σ2I)−Ξ  (1)
In Equation (1), W is a tap matrix expressed by a complex square matrix of an FFT block size using a tap coefficient as an element. Ξ is a complex diagonal matrix in which a frequency response of a propagation channel is arranged in a diagonal component. I is a unit matrix in which only the diagonal matrix is 1 and the remaining elements are 0. σ2 is a real number indicating a variance of thermal noise. In general, the tap matrix is expressed by the following equation when a kth transmitted signal within the FFT block is estimated by equalization.sest(k)=FHwH(k)R 
In this regard, F is a K×K complex square matrix to apply a DFT process. K is a block size of the FFT block. w(k) is a K×1 complex column vector for detecting the kth transmitted signal within the FFT block. R denotes a K×1 complex received signal vector of a frequency axis.
The tap matrix of Equation (1) is extended to simultaneously detect all symbols within the block and is extended as in the following equation.W=[w(1)w(2) . . . w(K)]
On the other hand, adaptive modulation and coding schemes are attracting attention as a technique of improving the efficiency of transmission. The adaptive modulation and coding schemes are a technique of selecting a modulation scheme which maximizes a transmission rate in a state in which the required quality is satisfied or selecting a coding rate in communication channel coding so as to maintain the equality of transmission for a temporal change of characteristics of a propagation channel.
In general, the adaptive modulation and coding schemes measure the quality of reception including a distortion of the propagation channel for each transmission opportunity based on error-rate characteristics measured under an AWGN (Additive White Gaussian Noise) environment without a distortion by the propagation channel. A modulation scheme or coding rate capable of accomplishing a maximum transmission rate while satisfying the required quality is determined.
For example, as combinations of available modulation schemes and coding rates, received SNRs necessary at a bit error rate 10−5 in BPSK of a coding rate ½, BPSK of a coding rate ⅔, QPSK of a coding rate ½, QPSK of a coding rate ¾, and QPSK of a coding rate ⅚ under an AWGN environment are set to 1 dB, 2 dB, 3 dB, 6 dB, and 8 dB, respectively. It is assumed that a received SNR measured in a certain transmission opportunity is 7.5 dB. In this case, the selected modulation scheme and coding rate satisfy the required quality at 7.5 dB or less, and QPSK of a coding rate ¾ in which a highest transmission rate may be achieved is set as the modulation scheme and the coding rate.
Here, in the case where the adaptive modulation and coding schemes are adopted in a single carrier scheme even though an equalization scheme called MMSE-FDE is used, a modulation scheme is generally adopted based on an SNR (Signal to Noise power Ratio) before equalization.    Non-Patent Document 1: D. Falconer, S. L. Ariyavisitakul, A. Benyamin-Seeyar, and B. Eidson, “Frequency Domain Equalization for Single-Carrier Broadband Wireless Systems,” IEEE Communications Magazine, vol. 40, pp. 58 to 66, April 2002.